jeh@crash.cts.com (Jamie Hanrahan) (01/29/88)
In article <2504@emory.uucp> osdbm@emory.uucp (Brendan Moriarity {EUCC}) writes: > ... it would be useful to break down the actual processor time spent > in each mode for the particular process being benchmarked. I agree, it certainly *would* be useful! Alas, VMS doesn't keep track of things to this level of detail. The DEC layered product called SPM might, but as we don't have it here I don't know. It's certainly possible to write a little hardware clock intercept routine that will track this information for you, but only if you're comfortable with writing elevated-IPL code. If anyone is interested, let me know, and I'll give you an outline.